Abstract
Design of time-frequency distributions (Tfds) that are robust to the impulse noise influence is considered. The robustTfds based on the robust short-time Fourier transform (Stft) are proposed. An efficient procedure to evaluate the robustStft is given. RobustTfds based on the robustStft have better energy concentration around the signal instantaneous frequency (If) than the robustStft itself. Also, theseTfds are more resistant to higher impulse noise than the robustTfds obtained using the local autocorrelation function (Laf) based minimization problem.
Résumé
L’article présente la conception de distributions temps-fréquence robustes en présence de bruit impulsif. Elles sont basées sur la transformée de Fourier à court terme robuste. Une procédure efficace permet de les évaluer. Leur concentration en énergie autour de la fréquence instantanée du signal est meilleure que celle de la transformée de Fourier elle-même. Ces distributions sont aussi plus résistantes à un fort bruit impulsif que celles qui sont obtenues par une minimisation fondée sur une fonction d’autocorrélation locale.
Similar content being viewed by others
References
Hlawatsch (F.),Boudreaux-Bartels (G. F.), Linear and quadratic time-frequency signal representations,Ieee Sig. Proc. Mag., Apr. 1992, pp. 21–67.
Cohen (L.), Time-frequency analysis, Prentice-Hall, 1995.
Hearon (S. B.), Amin (M. G.), Minimum-variance time-frequency distribution kernels,Ieee Trans. Sig. Proc.,45, no 5, May 1995, pp. 1258–1262.
Kwok (H. K.), Jones (D. L.), Improved instantaneous frequency estimation using an adaptive short-time Fourier transform,Ieee Trans. Sig. Proc.,48, no 10, Oct. 2000, pp. 2964–2972.
Boashash (B.), Estimating and interpreting the instantaneous frequency of a signal — Part I: Fundamentals, Proc.Ieee,80, No.4, Apr. 1992, pp. 521–538.
Flandrin (P.),Martin (W.), The Wigner-Ville spectrum of nonstationary random signals, in The Wigner distribution — Theory and applications in Signal Processing, W. Mecklenbrauker and F. Hlawastch, eds,Elsevier, 1997.
Middleton (D.), Non-Gaussian noise models in signal processing for telecommunications: New methods and results for class A and class B noise models,Ieee Trans. Inf. Th.,45, no 3, May 1999, pp. 1129–1150.
Katkovnik (V.), Robust M-periodogram,Ieee Trans. Sig. Proc.,46, no 11, Nov. 1998, pp. 3104–3109.
Djurović (I.), Katkovnik (V.), Stanković (L.J.), Median filter based realizations of the robust time-frequency distributions,Sig. Proc.,81, no 7, July 2001, pp. 1771–1776.
Djurovic (I.), Stanković (L.J.), Böhme (J. F.), Robust L-estimate based forms of the signal transforms and TF representations,Ieee Trans. Sig. Proc.,51, no 7, July 2003, pp. 1753–1761.
Huber (P. J.), Robust statistics,John Wiley & Sons, 1981.
Stankovic (L.J.), A method for time-frequency signal analysis,Ieee Trans. Sig. Proc,42, no 1, Jan. 1994, pp. 225–229.
Stankoviić (L.J.), A method for improved distribution concentration in the time-frequency signal analysis using the L-Wigner distribution,Ieee Trans. Sig. Proc,43, no 5, May 1995, pp. 1262–1269.
Stanković (L.J.), On the realization of the polynomial Wigner-Ville distribution for multicomponent signals,Ieee Sig. Proc. Let.,5, no 7, July 1998, pp. 157–159.
Barkat (B.),Stanković (L.J.), RobustPwvdfor analysis of polynomial fm signals in non-Gaussian noise, in Proc ofIeee icecs, Sep. 2002, pp. 1007–1010.
Djurović (I.), Stanković (L.J.): Realization of the robust filters in the frequency domain,Ieee Sig. Proc. Let.,9, no 10, Oct. 2002, pp.333–335
Cunningham (G. S.), Williams (W. J.), Kernel decomposition of time-frequency distributions,Ieee Trans. Sig. Proc.,42, no 6, June 1994, pp. 1425–1442.
Vakman (D.), On the analytic signal, the Teager-Kaiser energy algorithm, and other methods for defining amplitude and freqyency,Ieee Tram. Sig. Proc.,44, no 4, Apr.1996, pp.791–797.
Picinbono (B.), On instantaneous amplitude and phase of signals,Ieee Trams. Sig. Proc.,45, no 3, Mar. 1997, pp. 552–560.
Djurović (I.), Stanković (L.J.), Robust Wigner distribution with application to instantaneous frequency estimation,Ieee Trams,. Sig. Proc.,49, no 12, Dec. 2001, pp. 2985–2993.
Djurović (I.), Stanković (L.J.), Ohsumi (A.), Ijima (H.), Recursive realization of robustStft, in Proc. ofIeee icassp’2003,1, May 2003, pp. 157–160.
Pitas (I.),Venetsanopoulos (A. N.), Nonlinear digital filters: Principles and applications,Kluwer, 1990.
Kalluri (S.), Arce (G. R.), Robust frequency-selective filtering using weighted myriad filters admitting real-valued weights,Ieee Trams. Sig. Proc.,49, no 11, Nov. 2001, pp. 2721–2733.
Katkovnik (V.), Robust M-estimates of the frequency and amplitude of a complex-valued harmonic,Sig. Proc.,77, no 1, 1999, pp. 71–84.
Katkovnik (V.), Stanković (L.J.), Instantaneous frequency estimation using the Wigner distribution with varying and data driven window length,Ieee Trams. Sig. Proc.,46, no 9, Sep. 1998, pp. 2315–2325.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Djurovic, I., Stanković, L. & Barkat, B. Robust time-frequency distributions based on the robust short time fourier transform. Ann. Télécommun. 60, 681–697 (2005). https://doi.org/10.1007/BF03219942
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF03219942
Key words
- Signal theory
- Frequency time representation
- Impulsive noise
- Fourier transformation
- System robustness
- Statistical analysis